Pentium 4

[Gerhard Prade]

Hello,

i am not a developer, but have a question. I tried last time 
Gentoo-Linux and i think it is great to optimize a hole distribution of 
linux for a CPU-Achitcture (like Pentium 4 and not only for 386).
But Gentoo is for me not easy to use. So i want to use in future fedora, 
like i do now.
Now my question. It is posible to optimize one of the next releases of 
fedora for 686 CPU or greater like Pentium 4?

Thanks all, Gerhard

PS: Sorry for my bad english.

[Russell Coker]

On Tue, 3 Feb 2004 12:38, Gerhard Prade <gerhard.prade(a)uni-bielefeld.de&gt; 
wrote:
>  i am not a developer, but have a question. I tried last time
>  Gentoo-Linux and i think it is great to optimize a hole distribution of
>  linux for a CPU-Achitcture (like Pentium 4 and not only for 386).
>  But Gentoo is for me not easy to use. So i want to use in future fedora,
>  like i do now.
>  Now my question. It is posible to optimize one of the next releases of
>  fedora for 686 CPU or greater like Pentium 4? 
The level of optimisation that Gentoo does is not desirable (IMHO).

It means that every possible compiler bug related to CPU optimisation will be 
tickled, and that not all combinations will receive good testing which means 
that the chances of trying a random program on a random CPU is more likely to 
give bad results.

When reporting a bug it may be that you have an uncommon CPU and the 
maintainer of the package may find it difficult to get access to the same 
CPU.  This makes it more difficult to reproduce bugs.

Distributing RPMs for all the different CPUs takes significant archive space 
and significant numbers of CDs.

There have been many discussions of various ways of optimising programs.  The 
conclusion of the last discussion I saw was that there is no set of 
optimisation options that will be best for every program on a given CPU.  The 
question of when to use -Os and when to use -O2 is not an easy one to answer!

Finally, in most cases of compiler optimisation the payoff is small.  Changing 
the algorithms used by programs can deliver much more significant results, 
here are a few examples:

KDE performs poorly when large amounts of text are in a window, EG when kmail 
displays a message with a few megs of text.  I've seen the same operations 
run faster on other GUI software on 386 machines than kmail runs on P3 
machines!

kview and konqueror cause the X server to take up large amounts (600M or more) 
of memory when displaying 10,000 x 10,000 PNG files.  Mozilla and xv display 
the same files without requiring any significant resource usage from the X 
server and the entire system runs a lot faster.

As of my last test OpenLDAP was CPU bound for most of the process of importing 
an LDIF file and used only a single thread.  So for a medium size directory 
40M of CPU time would take 45M of clock time on a 2 CPU hyper-threaded 
machine, while ideally the work could have been done in 25M of clock time or 
less.

GPG will in some situations take ages to update the trustdb.  When I 
regenerated some of it's caches to try and track down another bug the 
update-trustdb time went from 23M to 1.5M!  This should be fixed so that it 
will always run fast for everyone.


If you make such an optimisation to one of the programs that is important to 
you then the results will be better than anything you could achieve through 
compiler optimisations.

Maybe we should create a list of programs who's performance is an issue for 
Fedora so that interested people can start investigating them?

-- 
http://www.coker.com.au/selinux/   My NSA Security Enhanced Linux packages
http://www.coker.com.au/bonnie++/  Bonnie++ hard drive benchmark
http://www.coker.com.au/postal/    Postal SMTP/POP benchmark
http://www.coker.com.au/~russell/  My home page

[Alan Cox]

On Tue, Feb 03, 2004 at 02:14:05PM +1100, Russell Coker wrote:
>  KDE performs poorly when large amounts of text are in a window, EG
> when kmail 
>  displays a message with a few megs of text.  I've seen the same operations 
>  run faster on other GUI software on 386 machines than kmail runs on P3 
>  machines! 
This is an X server problem in part. You also need to be sure to compare what
the apps are doing - there is a cost to rendering antialiased text for 
zillions of languages for example. If you switch your X server to unaccelerated
and turn on shadowfb your fonts probably go a lot quicker at the moment btw.

(Software render implementation has some design issues, XFree86 DDX lacks
 DMA 'get rectangle', and most servers still lack hardware render acceleration)


>  Maybe we should create a list of programs who's performance is an
> issue for 
>  Fedora so that interested people can start investigating them? 
Like "why does glibc's internationalisation post install script use 120Mb
?"

I think that is my favourite right now.

[Vincent]

On Tue, 03 Feb 2004 02:38:47 +0100
Gerhard Prade <gerhard.prade(a)uni-bielefeld.de&gt; wrote:

>  Hello,
>  
>  i am not a developer, but have a question. I tried last time 
>  Gentoo-Linux and i think it is great to optimize a hole distribution of 
>  linux for a CPU-Achitcture (like Pentium 4 and not only for 386).
>  But Gentoo is for me not easy to use. So i want to use in future fedora, 
>  like i do now.
>  Now my question. It is posible to optimize one of the next releases of 
>  fedora for 686 CPU or greater like Pentium 4?
>  
>  Thanks all, Gerhard
>  
>  PS: Sorry for my bad english.
>  
>  
>  -- 
>  fedora-devel-list mailing list
>  fedora-devel-list(a)redhat.com
>  http://www.redhat.com/mailman/listinfo/fedora-devel-list 
This has been discussed several times, A quick search pulls up:
https://listman.redhat.com/archives/shrike-list/2003-June/msg00686.html

In short it says the benifit is slight or unnoticable with the exception
of a few packages. kernel, glibc and SSL. If it is that big of a deal to you
options can be set in ~/.rpmrc and you can rebuild those 1 or two packages
yourself with little problem.

[Alan Cox]

On Tue, Feb 03, 2004 at 02:38:47AM +0100, Gerhard Prade wrote:
>  Now my question. It is posible to optimize one of the next releases
> of 
>  fedora for 686 CPU or greater like Pentium 4? 
It wouldn't run on other systems if the full optimisations were used.
Gentoo can do this as each user compiles the code and ends up with
binaries for their own CPU/setup that hopefully work.

Fedora does ship 686 optimised kernel, glibc and openssl, that seems to be
the key bits it helps. You can however rebuild any rpm source package with
other options (eg -Os for size optimisation or CPU specific optimisations)

[Alexander Dalloz]

Am Di, den 03.02.2004 schrieb Gerhard Prade um 02:38:
>  Hello,
>  
>  i am not a developer, but have a question. I tried last time 
>  Gentoo-Linux and i think it is great to optimize a hole distribution of 
>  linux for a CPU-Achitcture (like Pentium 4 and not only for 386).
>  But Gentoo is for me not easy to use. So i want to use in future fedora, 
>  like i do now.
>  Now my question. It is posible to optimize one of the next releases of 
>  fedora for 686 CPU or greater like Pentium 4?
>  
>  Thanks all, Gerhard
>  
>  PS: Sorry for my bad english. 
Hi Gerhard!

One of the main reasons Gentoo users claim for their system is the
performance boost by optimized compilations for their platform. But if
you look at the official Gentoo benchmark page - first official attempt
to make the subjective user's impression of performance gains countable,
other comparison charts are to be found using google - you will see that
the most significant improvement comes from prelinking. That is my
impression from the data on

http://www.gentoo.org/main/en/performance.xml

See yourself. The other speed values are on a level where I suspect you
will not feel any difference in general. Fedora uses prelinking.

Regards

Alexander


-- 
Alexander Dalloz | Enger, Germany | GPG key 1024D/ED695653 1999-07-13
Fedora GNU/Linux Core 1 (Yarrow) on Athlon CPU kernel 2.4.22-1.2149.nptl
Sirendipity 13:04:50 up 3 days, 12:05, load average: 0.12, 0.24, 0.20 
                   [ Γνωθι σ'αυτον - gnothi seauton ]

[Michael Schwendt]

On Tue, 03 Feb 2004 13:13:39 +0100, Alexander Dalloz wrote:

>  One of the main reasons Gentoo users claim for their system is the
>  performance boost by optimized compilations for their platform. But if
>  you look at the official Gentoo benchmark page - first official attempt
>  to make the subjective user's impression of performance gains countable,
>  other comparison charts are to be found using google - you will see that
>  the most significant improvement comes from prelinking. That is my
>  impression from the data on
>  
>  http://www.gentoo.org/main/en/performance.xml
>  
>  See yourself. The other speed values are on a level where I suspect you
>  will not feel any difference in general. Fedora uses prelinking. 
There are Gentoo users who are honest enough to admit that they don't
see any noticeable performance increase compared with Fedora Core 1.
Some even think that FC1 feels "snappier", and they assume it's due to
NPTL and/or kernel patches.

--